Article

Meta-analysis methods for synthesizing treatment effects in multisite studies: hierarchical linear modeling (HLM) perspective

---

Abstract

The objectives of the present mixed-effects meta-analytic application are to provide practical guidelines to: (a) Calculate treatment effect sizes from multiple sites; (b) Calculate the overall mean of the site effect sizes and their variances; (c) Model the heterogeneity in these site treatment effects as a function of site and program characteristics plus unexplained random error using Hierarchical Linear Modeling (HLM); (d) Improve the ability of multisite evaluators and policy makers to reach sound conclusions about the effectiveness of educational and social interventions based on multisite evaluations; and (e) Illustrate the proposed methodology by applying these methods to real multi-site research data. Accessed 58,759 times on https://pareonline.net from June 23, 2003 to December 31, 2019. For downloads from January 1, 2020 forward, please click on the PlumX Metrics link to the right.

Keywords: Meta Analysis, Research Methodology, Research Reports, Synthesis, Causal Models, HLM, Predictor Variables, Research Design, Statistical Inference

How to Cite:

Kalaian, S. A., (2002) “Meta-analysis methods for synthesizing treatment effects in multisite studies: hierarchical linear modeling (HLM) perspective”, Practical Assessment, Research, and Evaluation 8(1): 15. doi: https://doi.org/10.7275/hq9w-4881